top of page
Search

Thirty Years On, the Same Mistakes



Latham said it in 1994. Egan said it in 1998. The industry is still saying it in 2026. Three decades of construction reform have delivered little on cost. What, finally, is different now?

Every so often the industry rediscovers that it has a problem, commissions a report, agrees the diagnosis, and largely fails to change. The same conversation has now run for thirty years, and by some counts considerably longer Simon 1944 and Banwell 1960s.

Sir Michael Latham's Constructing the Team landed in 1994 and named the disease: an industry adversarial, fragmented and incapable of delivering for its clients. Four years later, Sir John Egan's Rethinking Construction pushed further, arguing that manufacturing had driven out cost and overrun by standardising and integrating, and asking why construction could not do the same.

The honest scorecard

Ten years on, Egan himself rated the industry's progress at around four out of ten. Andrew Wolstenholme's 2009 review put it more memorably: the industry could have had a revolution and achieved a bit of an improvement. The National Audit Office reckoned around thirty per cent in efficiency savings sat on the table. Most of it is still there.

The bigger culprit sits upstream

Bent Flyvbjerg's Oxford database, now covering more than sixteen thousand projects, describes an “iron law of megaprojects”: over budget, over time, under benefits, repeatedly. Around nine in ten megaprojects overrun, and only about half of one per cent land on budget, on time and on the promised benefits. The primary driver sits at the front end: optimism bias, procurement, and at times strategic misrepresentation, in the original business case.

HS2 illustrates the pattern in real time. A scheme first put at around thirty-three billion pounds is now forecast between roughly eighty eight and one hundred and three billion pounds, with government reviews attributing part of the rise to inflation never properly priced in, and the rest to missed scope and inefficient delivery. Assurance does not rescue a programme whose founding number was wrong before a spade went into the ground.

Yet Flyvbjerg's own remedy, reference class forecasting, only works with a trustworthy, comparable record of what past projects genuinely cost, line by line. That is precisely the record the industry mostly cannot produce, because its cost data was never assured in the first place. HS2's own reviews found budget control undermined by a consistent inability to produce reliable cost and schedule estimates. The front end and the monthly grind are the same problem, seen from opposite ends.

The layer the reports could not reach

Latham's report led to the 1996 Construction Act and the rise of NEC contracts, genuine gains that fixed the rules of engagement. What they could not fix was whether the cost data moving through those fairer processes was any good. Building information modelling gave the design world a single, assured source of truth. Nothing equivalent ever arrived for the commercial world. The drawing was digitised. How the building was paid for stayed in an earlier century.

Collaboration without a shared, trustworthy view of cost is just politeness.

What has changed in 2026 is that the data and tooling finally exist to assure the commercial layer the way modelling assured the design layer. This is not a fifth reform report. It is the missing instrument the earlier reforms kept pointing at and could not quite reach. Every generation of reformers believed it had found the missing piece. The difference this time is narrower and more mechanical: nobody needs converting to a philosophy, only a decision that the numbers must be made trustworthy, and the discipline to keep them that way.

 

This document was produced with AI assistance. In accordance with the Human Oversight requirements of this AIMS, all AI-generated content has been reviewed, validated and approved by the Director of Black Pear Advisory Ltd before issue.

Dr Martin Perks FRICS FQSi MICW MAC

Director, Black Pear Advisory Ltd

Worcester, UK  |  +44 7771 865271

 
 
 

Comments


bottom of page